1.1 Introduction

Anemia is a global public health concern particularly among women of reproductive age. It is characterized by a deficiency in red blood cells or hemoglobin levels. Inadequate access to healthcare, poor dietary practices, and various socioeconomic factors contribute to the prevalence of anemia. To address this critical health issue, the project explores into a comprehensive exploration of the risk factors associated with anemia among women of reproductive age using data from the Uganda Demographic and Health Survey (UDHS) of 2016.

The UDHS is a valuable repository of information, providing extensive insights into the health and well-being of Uganda’s population. In this project, we harnessed the power of data analysis and visualization to uncover key patterns that shed light on the prevalence of anemia among women aged 15 to 49 years. A diverse range of variables, including socioeconomic status and demographic characteristics were examined as the project aimed to paint a comprehensive picture of anemia’s risk factors. Through statistical analysis and data visualization techniques, the project attempted to explain the complex interplay of these factors.

2.1 Objectives of the project

The main objective of the project was to comprehensively analyze the multifaceted risk factors contributing to the prevalence of anemia in order to inform targeted interventions for its prevention and management.

This was achieved through these specific objectives:

  1. To establish the prevalence of anemia among women of the reproductive age

  2. To asses the prevalence of anemia among pregnant women and the influence of age

  3. To explore the urban-rural and regional differentials in anemia prevalence,

  4. To evaluate the relationship between household wealth, education level, and anemia risk

3.1 Project methodology

3.1.1 Packages installed and data loaded

Package installed included {here},{tidyverse},{janitor}.{here}, {gt} and {markdown}

3.1.2 Project Database

The dataset utilized in this project is a subset of the “Uganda Demographic Health Survey, data set” conducted by Uganda Bureau of Statistics (UBOS) in 2016. It encompasses a broad demographic spectrum, capturing data on women aged 15-49 years, children aged 0-59 months, and men aged 15-49 years within households selected for the male survey component. The survey incorporated four bio markers collected in the one-third of households selected for the male survey. In this project, results of one biomarker ** anemia testing** were used and in-relation to other related risk factors.To prepare the project data-set the UGIR7BFL.DTA was downloaded from the DHS website and a small data set ** udhs2016_raw** was read into R using the haven package, it has 18506 rows and 29 columns.

The data can he access here.

3.1.3 Data renaming

The data set required a renaming of its column variable names, and the rename function was employed to assign new names to these variables

4.1 Data wrangling and analysis

4.1.1 Data transformation

All labeled data was transformed into categorical factors using the as_factor function into a specific factor type that is compatible with the haven package. This step was crucial for ensuring that the data could be seamlessly integrated into the broader data analysis workflow. Additionally, to maintain the data set’s quality and suitability for further analysis, a data cleaning process was undertaken. Specifically, all rows with missing values in the ‘anemia level’ column were systematically eliminated using the filter function. This approach to data cleaning helps mitigate potential biases and inaccuracies, ensuring that the data set is robust and reliable for subsequent statistical and analytical procedures.

4.1.2 Data exploration and inspection

To explore the dataset comprehensively, numeric variables were summarized using measures of central tendency. Specifically, the mean, median, as well as the minimum and maximum values for each variable we provided in r. These statistics provided a valuable insight into the distribution and central values of the data, enbaling me to understand the typical values, potential outliers, and the overall range of each numeric variable. This summarization was a fundamental step in this data analysis, as it uncovered patterns, trends, and potential anomalies within the data set.

4.1.3 Measurement of outcome variables

Anemic and non-Anemic

The anemia level variable was re-coded into a binary outcome variable. Women aged 15 to 49 with severe, mild, or moderate anemia levels were categorized as ‘anemic,’ while those without any of these conditions were categorized as ’non-anemic.

Improved and unimproved water source for drinking

The water source variable was categorized into improved and unimproved. Improved water sources was redefined to include piped water, public taps, standpipes, tube wells, boreholes, protected dug wells and springs, and rainwater, and the rest were redefined as unimproved water source.

Improved and unimproved toilet facility

For a more precise understanding of sanitation conditions the toilet type variable was categorized into improved and unimproved facilities. ‘Improved’ facilities encompass flush/pour flush toilets to piped sewer systems, septic tanks, and pit latrines; ventilated improved pit (VIP) latrines; pit latrines with slabs; and composting toilets.

Computation for prevalence of anemia

5.1 Data visualization

Four plots were created each shedding light on the project objectives;

Plot 1

Plot 1 illustrates the prevalence of anemia among women aged 15 to 19 years. It is evident that 1 in every 3 women ( 33 percent), experienced some degree of anemia. Within this cohort, 26 percent had a mild form of anemia, indicating an ongoing health concern. Additionally, 6 percent faced the challenges of moderate anemia, while 1 percent battled with severe anemia.

These findings underscore the importance of targeted interventions and healthcare support for this vulnerable demographic.”

Plot 2

Plot 2 provides an insight into the prevalence of anemia among pregnant teenagers aged 15 to 19 years, a substantial 45 percent were found to be anemic, shedding light on the prevalence of anemia among this vulnerable demographic. The plot also paints a picture of anemia prevalence as women progress through various age groups. It further reveals that,as women advance in age, the prevalence of anemia steadily diminishes. However, an exception emerges among pregnant women aged 44 to 49 years, where the prevalence takes an unexpected upward turn.

Plot 3

Plot 3 unveils a spectrum of disparities that vary across regions, with figures ranging from a relatively modest 12% in urban Bukedi to a rather concerning 50% in the rural Acholi region. Notably, region-specific hotspots are presented in the plot that demand a focused attention. The rural Acholi region stands out with the highest prevalence at 51%, followed closely by urban Busoga at 47%, alongside rural West Nile at 44%, and urban Lango at 43%. These identified hotspots warrant tailored strategies to address the specific challenges contributing to elevated anemia rates.

Plot 4

Plot 4 illustrates the prevalence of anemia among women of reproductive age (15-49 years) in relation to household wealth and the source of drinking water. It highlights the disparities that exist within these categories. It reveals that women from lower wealth quintiles and those who used unimproved water sources for their drinking water were moderately anemic. Notably, the correlation between lower wealth quintiles and the prevalence of anemia, indicating a higher risk among women with limited economic resources. Additionally, the reliance on unimproved water sources for drinking water is associated with a higher likelihood of moderate anemia, revealing the critical importance of safe and clean drinking water access

Conclusion

In conclusion, the analysis of the selected subset of the 2016 Uganda Demographic and Health Survey has provided a multifaceted understanding of anemia prevalence among women of reproductive age, residence and regional backgrounds. i.e Woman’s age, urban-rural and region differentials, wealth index and source of water for drinking were associated with anemia prevalence. To understand the specific reasons for the observed changes, its recommended to conduct further analysis and consider additional factors that could explain the patterns observed.